ID #1016 Discovery and quantitation of metabolite biomarkers of ependymoma tumour relapse in cerebrospinal fluid
Madhumita Dandapani, Laudina Amugi, Chris Halsey, Richard Grundy, Dong-Hyun Kim, Alison WhitbyAbstract
Introduction
Ependymomas (EPN) have poor outcomes due to a 50% relapse rate. Most relapses occur within two years, suggesting persistent minimal residual disease (MRD) below the imaging detection threshold at the end of treatment. Detecting and tracking aberrant cancer-specific metabolites in cerebrospinal fluid (CSF) using a sensitive, minimally invasive liquid biopsy is promising as an MRD test, given the proximity of the tumour to the CSF space, and could help improve outcomes.
Methods
Liquid chromatography with tandem mass spectrometry (LC-MS/MS) of 25 µl CSF extracted with 75 µl methanol was used for untargeted metabolomics of EPN (n = 32; 14 days post-surgery; n = 10 at relapse) and control (n = 49) CSF. Subsequent targeted metabolomics of 10 metabolites with high linearity (R2>0.997), precision (CV ≤ 11%) and selectivity (Rs>1.1) was performed.
Results
Twenty metabolites were differentially abundant between EPN and controls. Seventeen metabolites were increased in EPN, including amino acids and derivatives (n = 9), one-carbon pathway metabolites (n = 1), carnitine cycle (n = 2), aerobic glycolysis (n = 2), purine catabolism (n = 2)and creatinine synthesis (n = 2). The betaine/creatinine ratio gave an area under the curve (AUC) of 0.93 (excellent) in receiver operating curve analysis (cut-off ratio=0.206; sensitivity of 98% and specificity of 78%). Targeted metabolomics confirmed the significantly increased concentrations of 10 metabolites (p < 0.05); four were excellent biomarkers of EPN with AUC >0.9. The majority of identified biomarkers correlated with published magnetic resonance spectroscopy and transcriptomic data from primary EPN tumours, as well as tumour bulk metabolomic data, and were conserved at relapse.
Conclusion:
This is the first CSF metabolomics study in EPN to detect residual brain tumour post-surgery with high sensitivity and specificity. Further validation is ongoing to develop this into a liquid biopsy assay for MRD monitoring and early relapse detection.